Have you ever wondered what the internet will look like in the future? When technologies have changed so much that the whole web is truly decentralized, all processes happen simultaneously everywhere. The loss of a node or outside interference can no longer affect what’s going on inside. Where data is available to all without censorship or restrictions, and is constantly accumulating.

A key difference between the internet we know now and its future incarnation will be this: AI will analyze and sift all data like anti-scientific information, baseless speculation, perhaps even brilliant thoughts. For those interested, it will provide footnotes to real research and publications. But besides simply showing such research, AI will also collect information from other sources: all rebuttals, criticism, related disciplines and studies, and analysis of other people’s opinions. There will be a continuous data analysis process, taking into account new data generated all the time, and all this will run in parallel and non-stop.

The exchange of ideas will be unlimited. In addition, the end user will not need to consider the truth value of statements: all this will be done automatically based on thousands of inputs; all data will be at your fingertips for you to check, if necessary. Having a database of relevant and purely scientific knowledge will make it possible to get the correct answer to almost any question, both the simple ones and those requiring extensive research. Because the main advantage of AI is the ability to predict. Armed with such a training database, we can dramatically improve people’s education, as well as create a universal self-updating scientific reference book.

If one adds to this concept a neural network able to interpret all knowledge in a human-understandable language, complete with relevant analogies, and even generate graphs, videos, and 3D visualization in real time, if there are no such referents in the database, then it will be possible to explain anything to anyone, regardless of what knowledge they had before.

Such a neural network is not a flight of imagination; many labs are already studying brain activity to create models that will help to understand what we’re thinking about. By analyzing this information, the neural network interpreter will be able to select the right approach for the individual, reminiscent of the cold-reading techniques used by psychics to obtain extra details when the flow of information changes according to their subject’s reaction. Only the AI will be literally inside the person’s head and get far more feedback during training. That’s what I would call the perfect teacher.

Now imagine a person walking along, wearing glasses and a cap. Inside the cap are non-invasive electrodes that read brain activity, allowing the wearer to access any information at will, which gets displayed on the glasses and is probably a good deal more relevant than any search results. If I spoke Latin, I would call it deus ex machina, literally “God from the machine” — a plot device whereby a complex situation is miraculously resolved (I don’t wish to hurt anyone’s feelings, but simply liken the power of this technology to the concept of God). It’s a pity that for the moment we’ll most likely lack the technological progress and computing power to bring this idea to life. But such technology can also be applied to speed up development in the right areas. Instead of a team of scientists with fragmented knowledge, we will have a self-generating scientific database that draws conclusions from the well of all human knowledge.